English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Volatility forecasts play a central role among equity risk measures. Besidestraditional statistical models, modern forecasting techniques based on machinelearning can be employed when treating volatility as a univariate, dailytime-series. Moreover, econometric studies have shown that increasing thenumber of daily observations with high-frequency intraday data helps to improvevolatility predictions. In this work, we propose DeepVol, a model based onDilated Causal Convolutions that uses high-frequency data to forecast day-aheadvolatility. Our empirical findings demonstrate that dilated convolutionalfilters are highly effective at extracting relevant information from intradayfinancial time-series, proving that this architecture can effectively leveragepredictive information present in high-frequency data that would otherwise belost if realised measures were precomputed. Simultaneously, dilatedconvolutional filters trained with intraday high-frequency data help us avoidthe limitations of models that use daily data, such as model misspecificationor manually designed handcrafted features, whose devise involves optimising thetrade-off between accuracy and computational efficiency and makes models proneto lack of adaptation into changing circumstances. In our analysis, we use twoyears of intraday data from NASDAQ-100 to evaluate the performance of DeepVol.Our empirical results suggest that the proposed deep learning-based approacheffectively learns global features from high-frequency data, resulting in moreaccurate predictions compared to traditional methodologies and producing moreaccurate risk measures.

DeepVol: Volatility Forecasting from High-Frequency Data with Dilated Causal Convolutions